/usr/local/lib64/python3.6/site-packages/torch/optim/__pycache__
Edit: /usr/local/lib64/python3.6/site-packages/torch/optim/__pycache__/nadam.cpython-36.pyc (5727B)
3
Eg, @ s4 d dl Z ddlmZ ddlmZ G dd deZdS ) N ) _functional) Optimizerc s4 e Zd ZdZd fdd Zej dd d Z ZS )NAdamaq Implements NAdam algorithm.
.. math::
\begin{aligned}
&\rule{110mm}{0.4pt} \\
&\textbf{input} : \gamma_t \text{ (lr)}, \: \beta_1,\beta_2 \text{ (betas)},
\: \theta_0 \text{ (params)}, \: f(\theta) \text{ (objective)} \\
&\hspace{13mm} \: \lambda \text{ (weight decay)}, \:\psi \text{ (momentum decay)} \\
&\textbf{initialize} : m_0 \leftarrow 0 \text{ ( first moment)},
v_0 \leftarrow 0 \text{ ( second moment)} \\[-1.ex]
&\rule{110mm}{0.4pt} \\
&\textbf{for} \: t=1 \: \textbf{to} \: \ldots \: \textbf{do} \\
&\hspace{5mm}g_t \leftarrow \nabla_{\theta} f_t (\theta_{t-1}) \\
&\hspace{5mm}if \: \lambda \neq 0 \\
&\hspace{10mm} g_t \leftarrow g_t + \lambda \theta_{t-1} \\
&\hspace{5mm} \mu_t \leftarrow \beta_1 \big(1 - \frac{1}{2} 0.96^{t \psi} \big) \\
&\hspace{5mm} \mu_{t+1} \leftarrow \beta_1 \big(1 - \frac{1}{2} 0.96^{(t+1)\psi}\big)\\
&\hspace{5mm}m_t \leftarrow \beta_1 m_{t-1} + (1 - \beta_1) g_t \\
&\hspace{5mm}v_t \leftarrow \beta_2 v_{t-1} + (1-\beta_2) g^2_t \\
&\hspace{5mm}\widehat{m_t} \leftarrow \mu_{t+1} m_t/(1-\prod_{i=1}^{t+1}\mu_i)\\[-1.ex]
& \hspace{11mm} + (1-\mu_t) g_t /(1-\prod_{i=1}^{t} \mu_{i}) \\
&\hspace{5mm}\widehat{v_t} \leftarrow v_t/\big(1-\beta_2^t \big) \\
&\hspace{5mm}\theta_t \leftarrow \theta_{t-1} - \gamma \widehat{m_t}/
\big(\sqrt{\widehat{v_t}} + \epsilon \big) \\
&\rule{110mm}{0.4pt} \\[-1.ex]
&\bf{return} \: \theta_t \\[-1.ex]
&\rule{110mm}{0.4pt} \\[-1.ex]
\end{aligned}
For further details regarding the algorithm we refer to `Incorporating Nesterov Momentum into Adam`_.
Args:
params (iterable): iterable of parameters to optimize or dicts defining
parameter groups
lr (float, optional): learning rate (default: 2e-3)
betas (Tuple[float, float], optional): coefficients used for computing
running averages of gradient and its square (default: (0.9, 0.999))
eps (float, optional): term added to the denominator to improve
numerical stability (default: 1e-8)
weight_decay (float, optional): weight decay (L2 penalty) (default: 0)
momentum_decay (float, optional): momentum momentum_decay (default: 4e-3)
.. _Incorporating Nesterov Momentum into Adam:
https://openreview.net/forum?id=OM0jvwB8jIp57ZJjtNEZ
Mb`??+?:0yE>r Mbp?c s d|kst dj|d|ks,t dj|d|d koBdk n sZt dj|d d|d kopdk n st dj|d d|kst d j|d|kst d
j|t|||||d }tt| j|| d S ) Ng zInvalid learning rate: {}zInvalid epsilon value: {}r g ?z%Invalid beta parameter at index 0: {}r z%Invalid beta parameter at index 1: {}z Invalid weight_decay value: {}z Invalid momentum_decay value: {})lrbetaseps weight_decaymomentum_decay)
ValueErrorformatdictsuperr __init__)selfparamsr r r
r r defaults) __class__ =/usr/local/lib64/python3.6/site-packages/torch/optim/nadam.pyr 5 s
zNAdam.__init__Nc C s d}|dk r&t j | }W dQ R X x| jD ]}g }g }g }g }g }g } |d \}
} x|d D ]} | jdk rd|j| | jjrtd|j| j | j| }
t|
dkrd|
d< d|
d< t j | t j
d |
d
< t j | t j
d |
d < |j|
d
|j|
d |j|
d |
d d 7 < | j|
d qdW t j |||||| |
| |d
|d |d |d d xNt
||D ]@\} }| j| }
|
d |
ddd|
d |d |
d< qpW q0W |S )zPerforms a single optimization step.
Args:
closure (callable, optional): A closure that reevaluates the model
and returns the loss.
Nr r z'NAdam does not support sparse gradientsr stepg ?
mu_product)Z
memory_formatZexp_avgZ
exp_avg_sqr r r r r
)beta1beta2r r r r
g ?g Q?)torchZ enable_gradZ param_groupsZgradappendZ is_sparse RuntimeErrorstatelenZ
zeros_likeZpreserve_formatFZnadamzip)r ZclosureZlossgroupZparams_with_gradZgradsZexp_avgsZ exp_avg_sqsZ mu_productsZ state_stepsr r pr" r r r r r G s\
,z
NAdam.stepr r )r r( r r r
)N) __name__
__module__ __qualname____doc__r r Zno_gradr
__classcell__r r )r r r s
- r )r r r$ Z optimizerr r r r r r
s